Use case

Your characters on video, from one key.

Your app holds the conversation. Corny renders the clips: send her picks or a sentence and get the clip back on a signed link.

Read the docs Get a key
POST /v1/videos
curl https://api.corny.studio/v1/videos \
  -H "Authorization: Bearer $CORNY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "describe": "she laughs on a beach at golden hour",
    "seconds": 8,
    "aspect": "portrait"
  }'

# { "id": "job_…", "status": "queued" }

  • No GPUs to run

    The rendering happens on our side.

  • Her face holds

    Characters built in the studio keep their face in every job.

  • Plain HTTP

    One POST to start, one GET to collect.

Adding video to an AI companion app

Your app already holds the conversation. What users ask for next is to see the character they talk to, in motion, looking like herself every time. Building that yourself means running video models, keeping GPUs busy and solving face consistency, which is a product of its own. Corny gives you that part as an API, so your team keeps working on the conversation.

The rendering happens on our side. Your server sends a job, gets an id back at once, and collects the finished clip when it is ready. The API renders exactly what the studio renders, so the clips you test by hand in the studio are the clips your users get. There is no model to pick and no prompt to tune for each character.

Generate character videos with one API call

Build the character first. Design her in the studio and accept her look, and from then on every job that names her carries that face. Then send POST /v1/videos with catalogue picks for her outfit, pose and place, or a describe sentence in your own words, along with the length and the shape. Portrait suits a phone screen, wide suits a desktop chat.

The response is a job id. Poll GET /v1/jobs/:id until the clip is ready, then fetch it from the signed link and show it in the conversation. Clips run up to ten seconds with the sound of the scene. It is plain HTTP: one POST to start, one GET to collect, from any language your backend speaks.

Characters are fictional adults made with AI, and every job has to follow our content policy. Keep your own app rules in place on top, since your users are the ones asking for scenes. A simple pattern is to map what a user asks for onto a short list of approved picks, and send a describe sentence only when the picks do not cover it.

API pricing for companion video

The API is priced in the same credits as the studio, priced by length, size and quality. A longer, bigger or sharper take costs more. Because the price follows the length, size and quality you send, you can work out the cost of a feature before you ship it.

The API is open to every account. New accounts start with 50 credits to try it, and plans add credits every month, cheaper each on a bigger plan: Developer is $99.90 for 6,000. Make a key under API keys in the studio and send your first job.

Questions people ask

Do I need my own GPUs to add video to my app?

No. Corny renders the clips. Your server sends a job and collects the finished clip from a signed link.

Will a character keep the same face across jobs?

Yes. A character built in the studio keeps her accepted face in every job that names her.

Does Corny provide the chat or the voice?

No. Corny renders video. A voice for characters is on the roadmap.

How do I get the finished clip?

Poll GET /v1/jobs/:id with the id you got back from POST /v1/videos. When the job is done, the clip is on a signed link.

The features behind it

Render her from your code.

Make a key in the studio and send your first job in minutes.

Read the docs Get a key